Introduction to Data Mining and Business Intelligence Lecture 1/DMBI/IKI83403T/MTI/UI

Transcription

1 Introduction to Data Mining and Business Intelligence Lecture 1/DMBI/IKI83403T/MTI/UI Yudho Giri Sucahyo, Ph.D, CISA Faculty of Computer Science, University of Indonesia

2 Objectives Motivation: Why data mining? What is data mining? Understand the drivers for BI initiatives in modern organizations Understand the structure, components, and process of BI 2

3 Motivation: Why data mining? Data explosion problem: Automated data collection tools and mature database technology lead to tremendous amounts of data stored in databases, data warehouses and other information repositories. We are drowning in data, but starving for knowledge! Data Mining: Extraction of interesting knowledge (rules, regularities, patterns, constraints) from data in large databases [JH]. Analysis of the large quantities of data that are stored in computers [DO], Alternative names KDD, knowledge extraction, data archeology, information harvesting, business intelligence, etc. 3

4 Data Rich but Information Poor Databases are too big Data Mining can help discover knowledge 4

5 Evolution of Database Technology Data Collection, database creation, network DBMS Relational data model, relational DBMS implementation RDBMS, advanced data models (extended-relational, OO, etc.) and application-oriented DBMS (spatial, scientific, engineering, etc.) Data mining and data warehousing, multimedia databases, and Web technology 5

6 Potential Applications See TSBD lecture notes Data Mining See Chapter 1 of DO Retailing Banking Credit Card Management Insurance Telecommunications Telemarketing Human Resource Management 6

7 Data Mining Should Not be Used Blindly Data mining find regularities from history, but history is not the same as the future. Association does not dictate trend nor causality. Some abnormal data could be caused by human. 7

8 Another view of BI BI is a broad field and it is viewed differently by different people. Common agreement on major components: A centralized repository of data data warehouse An end-user set of tools to create reports and queries from data and information and to analyze the data, information, and reports business analytics To find non-obvious relationship among large amounts of data data mining, for text text mining, for web web mining Business Performance Management (BPM) to set goals as metrics and standards and monitoring and measuring performance by using the BI methodology. 8

9 Drivers of BI Organizations are being compelled to capture, understand, and harness their data to support decision making in order to improve business operations Business cycle times are now extremely compressed; faster, more informed, and better decision making is therefore a competitive imperative Managers need the right information at the right time and in the right place Case Study 1: BI success story at Toyota Motor Company (Chapter 1 ET pg. 4-6). 9

10 Business Value of BI 10

11 Data Mining Functionality Association 11 From association, correlation, to causality Finding rules like A -> B Classification and Prediction Classify data based on the values ina classifying attribute Predict some unknown or missing attribute values based on other information Cluster analysis Group data to form new classes, e.g., cluster houses to find distribution patterns Outlier and exception data analysis Time series analysis (trend and deviation) Trend and deviation analysis: regression, sequential pattern, similiar sequences e.g. Stock analysis

12 Sarbanes-Oxley Act of 2002 (extracted from Gartner, Inc., 2004) The Sarbanes-Oxley Act of 2002 mandates drove one firm to implement a new financial performance management system, capable of meeting the new requirements to: Perform flawless analysis and compilation of thousands of transactions and journal entries. Balance more access to data with the need to control access to sensitive insider information. Deliver reports to the SEC in less time. 12

13 Sarbanes-Oxley Act of 2002 (extracted from Gartner, Inc., 2004)... continued Within the overarching goal of achieving financial-reporting compliance, these objectives included the following: Get more eyes on the data and KPI and build in strict security controls Provide live reports that allow people to drill down to the lowest level of transaction detail Proactively scour the financial databases for anomalies, using variance triggers Gather all financial data into a cohesive database Complement accounting and budgeting applications for flexible reporting, free-form investigation, and automated data analysis. BI can proactively alert specific individuals whenever an anomay is detected. 13

14 14 Now let us see some screenshots...

15 Dashboard 15

16 And another dashboard... 16

17 And another dashboard... 17

18 Financial Reporting 18

19 19 Back to theory...

20 KDD Process Interpretation/ Knowledge Evaluation Transformation Data Mining Preprocessin g Selection Target Data Data 20

21 Steps of a KDD Process Learning the application domain: 21 relevant prior knowledge and goals of application Creating a target data set: data selection Data cleaning and preprocessing: (may take 60% of effort!) Data reduction and projection: Find useful features, dimensionality/variable reduction, invariant representation. Choosing functions of data mining summarization, classification, regression, association, clustering. Choosing the mining algorithm(s) Data mining: search for patterns of interest Interpretation: analysis of results. visualization, transformation, removing redundant patterns, etc. Use of discovered knowledge

22 Teradata Advanced Analytics Methodology (similar to CRISP-DM) 22

23 Structure and Components of BI 23

24 Structure and Components of BI... continued Data Warehouse Data flows from operational systems (e.g., CRM, ERP) to a DW, which is a special database or repository of data that has been prepared to support decision-making applications ranging from those for simple reporting and querying to complex optimization Business Analytics/OLAP Software tools that allow users to create ondemand reports and queries and to conduct analysis of data 24

25 Structure and Components of BI... continued Data Mining Data mining is a class of database information analysis that looks for hidden patterns in a group of data that can be used to predict future behavior Used to replace or enhance human intelligence by scanning through massive storehouses of data to discover meaningful new correlations, patterns, and trends, by using pattern recognition technologies and advanced statistics 25

26 Structure and Components of BI... continued Business Performance Management (BPM) Based on the balanced scorecard methodology a framework for defining, implementing, and managing an enterprise s business strategy by linking objectives with factual measures Dashboards A visual presentation of critical data for executives to view. It allows executives to see hot spots in seconds and explore the situation 26

27 BI: Today and Tomorrow Recent industry analyst reports show that in the coming years, millions of people will use BI visual tools and analytics every day BI takes advantage of already developed and installed components of IT technologies, helping companies leverage their current IT investments and use valuable data stored in legacy and transactional systems Some Issues: Mining information from heterogeneous databases and global information systems Handling relational and complex types of data Efficiency and scalability of data mining algorithms 27